Bitcoin Forum
December 14, 2024, 08:09:21 PM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 [23] 24 25 26 27 28 29 »
  Print  
Author Topic: CoinWallet.eu Stress Test Cancelled + Bitcoin Giveaway  (Read 98921 times)
basil00
Member
**
Offline Offline

Activity: 60
Merit: 10


View Profile
September 13, 2015, 05:15:20 AM
 #441

Quote
@basil00, you can use predefined 'k' for signing ECDSA. This saves some space in blockchain.

Thanks for the tip -- was not aware of this.  I have to figure out how do this with cbitcoin though.

Note that if the replacement transaction is bigger the miner may prefer the original anyway (better fee rate).
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 05:28:17 AM
 #442

Note that if the replacement transaction is bigger the miner may prefer the original anyway (better fee rate).
You can vary the number of inputs of your tx.
Take several inputs from one tx and several from another, and doublespend two transactions!  Grin

In fact miners do not use replace-by-fee today.
basil00
Member
**
Offline Offline

Activity: 60
Merit: 10


View Profile
September 13, 2015, 06:24:48 AM
 #443

Quote
@basil00, you can use predefined 'k' for signing ECDSA. This saves some space in blockchain.

I tried but no luck so far, still get big sigs.  What exactly is the K-value required (kinv for ECDSA_sign_ex)?  Does it depend on the priv key?
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 06:51:02 AM
 #444

I tried but no luck so far, still get big sigs.  What exactly is the K-value required (kinv for ECDSA_sign_ex)?

My code is very dirty and I do not understand it myself Smiley
Code:
static BIGNUM* bn_kinv = BN_new ( );
...
static const MyKey32 k ( QByteArray::fromHex ( "7fffffffffffffffffffffffffffffff5d576e7357a4501ddfe92f46681b20a0" ));
BN_bin2bn ( k.constPtr ( ), 32, bn_kinv );
...
EC_POINT_mul ( group, tmp_point, bn_kinv, 0, 0, ctx );
EC_POINT_get_affine_coordinates_GFp ( group, tmp_point, bn_x, 0, ctx );
...

Does it depend on the priv key?
no
Evil-Knievel
Legendary
*
Offline Offline

Activity: 1260
Merit: 1168



View Profile
September 13, 2015, 09:35:21 AM
Last edit: April 17, 2016, 07:46:13 PM by Evil-Knievel
 #445

This message was too old and has been purged
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 10:29:03 AM
 #446

What exactly is the K-value required (kinv for ECDSA_sign_ex)?

Code:
static const MyKey32 k ( QByteArray::fromHex ( "7fffffffffffffffffffffffffffffff5d576e7357a4501ddfe92f46681b20a0" ));

BN_bin2bn ( k.constPtr ( ), 32, bn_kinv );              // convert bytearray to openSSL bignumber object
EC_POINT_mul ( group, tmp_point, bn_kinv, 0, 0, ctx );
EC_POINT_get_affine_coordinates_GFp ( group, tmp_point, bn_x, 0, ctx );

BN_nnmod ( retr, bn_x, bn_order, ctx );                 // retr is R-part of signature
BN_mod_inverse ( bn_kinv, bn_kinv, bn_order, ctx );     // todo: seems to me that this should be performed once!
BN_mod_mul ( tmp, ecprivkey, retr, bn_order, ctx );     // ecprivkey is my privkey as openSSL bignumber object
BN_mod_add_quick ( rets, tmp, BN_bin2bn ( dgst.constPtr ( ), 32, m ), bn_order );
BN_mod_mul ( rets, rets, bn_kinv, bn_order, ctx );      // rets is S-part of signature

// now encode pair of <retr,rets> to DER

amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 01:13:53 PM
 #447

I tried but no luck so far, still get big sigs.  What exactly is the K-value required (kinv for ECDSA_sign_ex)?
Congratulations!
https://blockchain.info/tx/f3aca5442d7e0129d1060d34406bdfa6bb052e60977313dd56b7d8a589f4af78
Now remove unused bytes from OP_RETURN to make your result perfect!  Grin
rz20
Legendary
*
Offline Offline

Activity: 1330
Merit: 1001


View Profile
September 13, 2015, 01:16:26 PM
 #448

I tried but no luck so far, still get big sigs.  What exactly is the K-value required (kinv for ECDSA_sign_ex)?
Congratulations!
https://blockchain.info/tx/f3aca5442d7e0129d1060d34406bdfa6bb052e60977313dd56b7d8a589f4af78
0 btc transacted. Damn this really doesn't work and the OP is just spamming the blockchain.
Evil-Knievel
Legendary
*
Offline Offline

Activity: 1260
Merit: 1168



View Profile
September 13, 2015, 01:21:04 PM
Last edit: April 17, 2016, 07:46:07 PM by Evil-Knievel
 #449

This message was too old and has been purged
yakuza699
Hero Member
*****
Offline Offline

Activity: 935
Merit: 1002


View Profile
September 13, 2015, 01:22:49 PM
 #450

I tried but no luck so far, still get big sigs.  What exactly is the K-value required (kinv for ECDSA_sign_ex)?
Congratulations!
https://blockchain.info/tx/f3aca5442d7e0129d1060d34406bdfa6bb052e60977313dd56b7d8a589f4af78
0 btc transacted. Damn this really doesn't work and the OP is just spamming the blockchain.
Wrong.Sending more than 0BTC to an OP_RETURN output would be non-standard.As you can see he gave everything as fees for the miners and you should know that OP_RETURN outputs does not increase UTXO.

▄▄▄▄▄▄▄▄
▄▄▄▄▄▄
▄▄▄▄
BTC BitDice.me 
.
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 01:23:56 PM
 #451

0 btc transacted. Damn this really doesn't work and the OP is just spamming the blockchain.
The OP is doing nothing. Everything you see here has been done by others.
This is some kind of cleenup the dust.
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 01:26:31 PM
 #452

The private keys however are entirely useless.
I took ~4BTC from these keys. What do you mean by "useless"? Useless for you?  Grin
basil00
Member
**
Offline Offline

Activity: 60
Merit: 10


View Profile
September 13, 2015, 01:29:03 PM
 #453

Congratulations!
https://blockchain.info/tx/f3aca5442d7e0129d1060d34406bdfa6bb052e60977313dd56b7d8a589f4af78
Now remove unused bytes from OP_RETURN to make your result perfect!  Grin

Thanks.  After fiddling around with openSSL and getting nowhere I switched to libsecp256k1 and got it working first try.  All you need to do is make the nonce function return the magic value for K.

The new txs are tagged 'S' or 'SS'.  I will remove this byte soon.

Some of the older encoded txs (XX, XY, XW, R, DS) will likely bounce around for a while.  I've set my script to double-spend them.

Also, only SIGHASH_ALL works, otherwise other nodes reject the transaction.  Non-standard or maybe I stuffed something up?
yakuza699
Hero Member
*****
Offline Offline

Activity: 935
Merit: 1002


View Profile
September 13, 2015, 01:32:18 PM
 #454

The fees for these small outputs are higher than the values in those outputs.
Yes that's what you need to do if you want your tx to get confirmed.
So here is what I think: The OP wanted to perform the stress test but realized that he will not be able to accumulate those dust outputs (which he before created by splitting 250 BTC up in small amounts) without spending another 250 BTC on transaction fees.
I think it's wrong.Maybe OP was just too lazy to do it himself so instead released the priv keys so that other people would do the "stress test" for him.And you know if OP would have gave me the keys I would have got at least 100BTC from it.
Now he realized that he essentially burned 250 BTC and the release of those private keys is one last desperate move to reestablish his good reputation (by appearing generous). The private keys however are entirely useless.
As amaclin said useless for who?

▄▄▄▄▄▄▄▄
▄▄▄▄▄▄
▄▄▄▄
BTC BitDice.me 
.
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 01:36:23 PM
Last edit: September 13, 2015, 02:04:41 PM by amaclin
 #455

Quote
Some of the older encoded txs (XX, XY, XW, R, DS) will likely bounce around for a while.  I've set my script to double-spend them.
No need to do it. Because my script converts everything to "my address".
And your script converts my result to OP_RETURN.
Unfortunately to me, I can not beat you Smiley

OK, the game is over. From this moment all dust outputs from compromised private keys will go to miners only.  Grin

@basil00, You took my profits, the son of a bitch !  Grin Grin Grin
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 01:53:54 PM
 #456

Sending more than 0BTC to an OP_RETURN output would be non-standard.
No, this is standard output. The amount will be burned
yakuza699
Hero Member
*****
Offline Offline

Activity: 935
Merit: 1002


View Profile
September 13, 2015, 03:12:23 PM
Last edit: September 13, 2015, 03:25:38 PM by yakuza699
 #457

Sending more than 0BTC to an OP_RETURN output would be non-standard.
No, this is standard output. The amount will be burned
Wrong. https://bitcoin.org/en/developer-guide#non-standard-transactions "As of Bitcoin Core 0.9.3 standard null data outputs must receive zero satoshis".Still think I am wrong?
Code:
010000000133c3d38bb4a30bd4595707e13ca17d948d051365b166f50da26527680f0bddf0000000008a47304402205a277704ca398dfdfee958ab6eb3af098d88b60c34712ff5bceab9401a715c4d02202523894f0924136bde8ae07b8b6cf06cb2e75c175172c31e6800db7d9adca648014104f64e3d965b8f4aa507665a7a765dab9a78192c1dc7d87743696ee7adc714de336b8435c3e21ae72d917fe1e588f54e632a8448611a8a1b20bbe9066c0ec78a75ffffffff0290512d00000000001976a91474824ef4fc87f5e6a9c7800ca5afd67e49dac34188ac0000000000000000066a045445535400000000
Good luck relaying it.

▄▄▄▄▄▄▄▄
▄▄▄▄▄▄
▄▄▄▄
BTC BitDice.me 
.
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 04:25:14 PM
 #458

Still think I am wrong?
Yes. I think you are wrong.
This is the actual code:

   BOOST_FOREACH(const CTxOut& txout, tx.vout) {
        if (!::IsStandard(txout.scriptPubKey, whichType)) {
            reason = "scriptpubkey";
            return false;
        }

        if (whichType == TX_NULL_DATA)
            nDataOut++;

        else if ((whichType == TX_MULTISIG) && (!fIsBareMultisigStd)) {
            reason = "bare-multisig";
            return false;
        } else if (txout.IsDust(::minRelayTxFee)) {
            reason = "dust";
            return false;
        }
    }

    // only one OP_RETURN txout is permitted
    if (nDataOut > 1) {
        reason = "multi-op-return";
        return false;
    }


The reference client either increases the total count of null-data outputs in tx or checks for dust limits.
But not both.
So, it is possible to burn any amount via OP_RETURN, even zero.

Note, that bc.i still treats null-data output as non-standard
yakuza699
Hero Member
*****
Offline Offline

Activity: 935
Merit: 1002


View Profile
September 13, 2015, 05:07:42 PM
 #459

Still think I am wrong?
Yes. I think you are wrong.
I am NOT saying that, that transaction is invalid.I am saying that, that transaction is non-standard.Try to relay the given tx and see yourself.

▄▄▄▄▄▄▄▄
▄▄▄▄▄▄
▄▄▄▄
BTC BitDice.me 
.
amaclin
Legendary
*
Offline Offline

Activity: 1260
Merit: 1019


View Profile
September 13, 2015, 05:37:38 PM
 #460

I am NOT saying that, that transaction is invalid.I am saying that, that transaction is non-standard.Try to relay the given tx and see yourself.
take it confirmed
https://tradeblock.com/bitcoin/tx/ba1ed5b4b036f9adddc4092273b19381b525698b7cf51a6fbe1112466ba3d085
Pages: « 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 [23] 24 25 26 27 28 29 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!